Telemedicine was a rapidly growing market even before the COVVI-19 pandemic, but recent events highlighted the need for accessible distance health care and accelerated the discovery of new solutions in space. One of these efforts is Kelvin Health, a Bulgarian company at an early stage that develops a mobile platform for AI thermal analysis which can allow users to detect and monitor respiratory, oncological and other thermal verifiable diseases although visible symptoms appear.
Started almost three months ago during the forty coronavirus, Kelvin is a spin-off of the Imagga image recognition company, which has to date analyzed and marked more than 15 billion photos for its customers. Based on this experience, in March, the team began to collaborate with medical researchers and developed a prototype of an application which, with the help of a thermal camera, allows the non -invasive discovery of changes and anomalies in the bloodstream.
Quick advance until June, the Kelvin application has been launched in beta version, there are planned pilot projects and technology marketing for certain use cases, and the company has been selected to participate Hub Bulgaria Innovation Program for acceleration on the American market.
Medicine and thermography meet artificial neural networks
Georgi Kadrev, CEO of Kelvin HealthShare that if there is a change in the distribution of blood, it immediately affects the thermal instantaneous of the body. It can take a few days for symptoms such as cough and fever are felt, however, signs of infection are already present.
At first glance, the detection of anomalies in blood heat and distribution does not look like a task that can be done at home, but Kelvin aims to change this so that more people can take measures in time and avoid complications. Following the advice of their general practitioner, of course.
So how does it all work? Well, first of all, the user must find a thermal camera. Kelvin is currently compatible with several of these cameras, but the company plans to associate with a supplier and provide its own equipment, more suitable for medical cases and three to four times cheaper than the devices available on the market.
With this camera connected to their phone, people can take thermal photos of themselves – more frequently, the more precise the results, but from Kelvin recommend that the threshold is at least four times a day. Then, the application maps the distribution in real time of heat and blood in the body and compares the visual collected with medical information in the system.
“”We use an image processing algorithm based on AI neural networks. Trained with a large number of images, a neural network can learn what the different markers are and what are the typical structures of these images. From there, any difference compared to the standard line can be easily visible. In our case, using Big Data sets of thermal photos of healthy and sick people, we can then identify anomalies and calculate the probabilities so that the user has viral infection or necrosis, as the case may be, “explains Georgi Kostadinov, CTO at Kelvin Health.
Kelvin collects data for his algorithms from a few different sources, the principal being medical staff who validated and classified images of interest. The algorithm itself does not only use data from thermal images, but also demographic data and generated by the user such as blood pressure. At the end, everything is combined into a “medical algorithm” to correctly detect the occurrence of an anomaly.
What is the next step
Kelvin is currently focusing on R&D, the first pilots and fundraising, the official launch planned for the moment for the last quarter of the year.
“We have analyzed the verticals and the use cases have the most sense to continue from a commercial and technical point of view and we have already focused our attention on some – one of them are respiratory infections, the second – oncological and the third – sports injuries and vascular disorders related to athletes,” explains Kadrev.
Overall, the next activities on the roadmap include large -scale clinical trials and dedication to regulatory compliance, massive data collection and improving AI accuracy and partnerships with thermal cameras suppliers.
Regarding marketing, the team has some potential commercial models in mind. Consumers will pay monthly costs to use the platform while on the B2B side, the company will sell licenses for the whole software battery to hospitals and health establishments so that doctors can monitor patients at home.
“We want to make technology accessible to everyone to transport in their pocket and be able to use it at home. For us, preventive medicine and the early diagnosis of diseases have great potential not only from the point of view of companies, but also as an impact on our health system and our well-being ”concludes Kadrev.